Terry O. Herndon

Magnetic Film Memory Design

Thin magnetic films of permalloy have characteristics ideal for high-speed digital storage. A simple rotational model modified to include the effects of wall switching and dispersion of the preferred direction of magnetization provides a basis for describing properties of engineering interest. A selection system has been chosen which allows great latitude in film uniformity. Production of films with magnetic properties uniform to within ± 10 per cent is readily achieved. Specifications for operation in a destructive mode can easily be met by existing film arrays; the nondestructive mode is considerably more stringent unless very small signals can be tolerated. The first film memory has been in reliable operation since the summer of 1959. It has 32 ten-bit words and has been operated with a minimum cycle time of 0.4 , μsec. Higher speed and larger capacities will require higher bit densities and improved techniques to minimize undesirable coupling between drive and sense lines. The use of 10 × 60 mil rectangles, balanced sense windings, and longer words will hopefully permit memories of about 200,000 bits with cycle time under 0.2 μsec.

Spatially constrained skin electroporation with sodium thiosulfate and urea creates transdermal microconduits

Controlled transport of molecules through the skins main barrier, the stratum corneum (SC), is a long standing goal of transdermal drug delivery. Traditional, needle-based injection provides delivery of almost any water soluble compound, by creating a single large aqueous pathway in the form of the hollow core of a needle, through which drug is delivered by pressure-driven flow. We extend previous work to show that SC-spanning microconduits (here with diameters of about 200 microm) can be created in vivo by skin electroporation and low-toxicity, keratolytic molecules (here sodium thiosulfate and urea). A single microconduit in isolated SC can support volumetric flow of the order of 0.01 ml s(-1) by a pressure difference of only 0.01 atm (about 10(2) Pa), demonstrating that the SC barrier has been essentially eliminated within this microscopic area.

Spatially constrained localized transport regions due to skin electroporation.

Rapid, controlled molecular transport across human skin is of great interest for transdermal drug delivery and minimally invasive chemical sensing. Short, high-voltage pulses have been shown previously to create localized transport regions in the skin. Here, we show that these regions can be constrained to occur at specific sites using electrically insulating masks that restrict the field lines. The increase in total ionic and molecular transport per area was comparable to the levels observed in unconstrained electroporation of human skin. Constraining the area of intervention to encompass small areas of interest, a primary feature in the design of microdevices for transdermal drug delivery, can provide the same levels of flux as the unconstrained case.

Process considerations in restructurable VLSI for wafer-scale integration

Wafer-scale integration has recently been demonstrated using a technique called Restructurable VLSI. An array of logic cells embedded in programmable interconnect is fabricated on the wafer. All the parts are tested by wafer probing, and links are made or broken with a laser to wire the complete system. One such chip, a digital integrator 24 cm2in area with 25 MHz input data rate, has been successfully programmed. This paper will describe the RVLSI concept and discuss several aspects of wafer fabrication which are unusual in this technology.

A progress report on large capacity magnetic film memory development

In 1964 we proposed an approach to magnetic film memory development aimed at providing large, high-speed, low-cost random-access memories. Almost without exception, all early attempts at film memory design emphasized speed with little consideration for the potential of batch-fabrication to reduce costs. Based on our earlier work in building the first film memory in 1959, and a 1,000 word, 400 nsec model for the TX-2 computer in 1962, we had reached some fundamental conclusions about the compatibility of high speed and low cost for destructive-readout film memories.

Recent Advances in Skin Electroporation: Mechanism and Efficacy

Rapid, controlled molecular transport across human skin is of great interest for transdermal drug delivery and noninvasive chemical sensing. The main barrier is the stratum corneum (SC), which can be described by a “brick wall” model in which the dead, hydrated corneocytes are the bricks, and the surrounding multilamellar lipid bilayer membranes are the mortar. Small lipid-soluble molecules can partition into the SC, and then diffuse across the lipid bilayer membranes, but water soluble molecules, particularly charged molecules, cannot penetrate significantly by this route. Pre-existing aqueous pathways associated with the skin’s appendages (sweat gland duets and hair follicles) admit water soluble molecules, and are believed to provide one major route for iontophoresis, in which low voltages are used to move ions and molecules across the skin. However, iontophoresis often provides molecular fluxes which are smaller than needed.

Multilevel interconnect planarization by voltage and laser programmable links using ion implantation

A technique is described whereby implantation of silicon through a mask into the intermetal insulator modifies the insulation. After deposition and definition of the upper metal, the implanted regions between metal levels act as voltage programmable links. Application of a voltage between upper and lower metal electrodes causes the implanted insulation to become conductive, producing a low-resistance, planar, vertical connection. Alternatively, these implanted areas can be rendered conducting by exposure to a focused laser beam.<<ETX>>

Multilevel Interconnect Structures

My intent is to review multilevel interconnect for I.C. fabrication; to look at why multilevel is essential; to assess the current status of multilevel interconnect with regard to density, number of levels, and processing problems; to estimate future needs for multilevel interconnect and consider some possible new ways to achieve these goals. First, to establish some assumptions, Figure 1 shows an interesting relationship with regard to capacitance. CL is the line to line capacitance and CS is the line to substrate capacitance on interconnect for integrated circuits. As the metal pitch or center to center spacing goes down, the lines get narrower and therefore, the line to substrate capacitance is reduced. But at the same time, the spacing between lines is reduced and the line-to-line capacitance goes up. The result is a minimum in these two capacitances at around a 2 μm pitch. That would mean 1 μm spaces between 1 μm wide conductors which might be 1 μm thick. The other factor not shown here is that as the pitch goes down, the resistance is starting to go up due to conductor narrowing, and the RC time constant increases. One of the factors in multilevel interconnect is that usage of silicon by active devices on the substrate, ranges from 4%-8% depending on the type of circuit and the densities. The speed of the devices are no longer the limiting factor at all. The RC time constant of the interconnect governs the total system speed. This means that the devices have to be put closer together, interconnect needs to be made denser, and at the same time reduce the RC time constant as much as possible. Therefore, I am assuming that a 2 μm metal pitch will be the reasonable expectation to be met in the future.